Our previous research has indicated that, adjusting for specialty, a few physicians account for the major part of the medical malpractice losses and a previous record of high loss is predictive of high future loss. It is not now known whether an adverse claims record is an indicator of poor physician performance or not. The requisite data on risk factors of patients of physicians with poor versus exemplary malpractice records are not available. Nor is it known whether physicians with exemplary records take more precautions to avoid suits. To learn more about why some physicians' claims record vary, we focus on births and birth-related outcomes. We shall interview mothers in 1991 in Florida who had babies in 1986 to gather information of types of care received, patient risk factors, satisfaction with care received, birth outcomes, and conditional on an adverse outcome, whether the mother filed a claim against her obstetrician. Patients of obstetricians with very high losses will be oversampled. Data on previous loss experience will come from a closed claims file maintained by the Florida Department of Insurance. These data will be supplemented with data from Florida vital statistics and hospital charts. For purposes of analysis, we consider four stages: (1) mother's choice of physician for obstetrical care; (2) prenatal period, focusing on utilization patterns; (3) labor and delivery, including choice of delivery method and complications; and (4) child's early years--health outcomes and decision to file a claim. The analytic work will use models of search behavior (for studying patient choice of physician for obstetrical care) and deterrence to study precautionary actions of physicians who were at varying risks of being sued. In the end, we should know much more than now about the birth outcomes, patient satisfaction, treatment styles, and patients of physicians with very adverse claims experience versus other physicians.